International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS) Volume VI, Issue III, March 2017 | ISSN 2278-2540 Compare Efficiency of Different Multipliers Using Verilog Simulation & Modify an Efficient Multiplier Abhishek Bhattacharjee1, Anindya Sen2 1,2 Department of Electronics and Communication Engineering (VLSI), Heritage Institute of Technology, Kolkata, West Bengal Abstract- Multipliers are the fundamental components in many data encoding, the parallelization or pipelining of operations, digital signal processing systems. Many important signal and the tuning of input bit-patterns to reduce switching processing systems are designed on VLSI platform as the activity. integration growing rapidly. The signal processing systems & applications requires large computational capability, hence takes A basic multiplier can be divided into three parts i) partial considerable amount of energy. In the VLSI system design product generation ii) partial product addition and iii) final performance, area and power consumption are three important addition [5]. parameters, of which power consumption is the gets prime importance. In today’s world, power consumption is very There are number of techniques that to perform binary important factor. The largest contribution to the power multiplication. In general, the choice is based upon factors consumption in multiplier is due to generation and reduction of partial products. So it is very much important to know the such as latency, throughput, area, and design complexity. efficiency of different multipliers. This paper represents a More efficient parallel approach uses some sort of array or detailed comparison between array multiplier, Wallace tree of full adders to sum partial products. Array multiplier, multiplier, Dadda multiplier, modified array multiplier on the booth multiplier, Wallace Tree multipliers are some of the basis of speed, area, power consumption using verilog standard approaches to have hardware implementation of simulation. binary multiplier which are suitable for VLSI implementation Keywords- Multiplier, Array, Wallace, Dadda, Modified Array, at CMOS level .This paper represents Array multiplier(using Verilog. ripple carry adder),Array multiplier(using ripple carry adder & carry save adder),Array multiplier(using carry save adder & I. INTRODUCTION carry look ahead adder) ,Baugh-wooley multiplier, Wallace tree, Dadda tree multiplier, Modified array multiplier and ultiplier is one of the basic functional unit in digital compare their efficiency. M signal processor (DSP). Most of the high performance DSP systems rely on hardware multiplication to achieve high II. OBJECTIVE data throughput [5]. and since, multiplication dominates the The objective of good multiplier to provide a physically execution time of most DSP algorithms, therefore high-speed compact high speed and low power consumption unit. Being a multiplier is much desired[8]. Currently, multiplication time is core part of arithmetic processing unit multipliers are in still the dominant factor in determining the instruction cycle extremely high demand on its speed and low power time of a DSP chip. consumption. Multiplication is a fundamental operation in most signal To find the most efficient timing characteristics and area processing algorithms. Multipliers have large area, long report generated by the Xilinx tool for various multipliers will latency and consume considerable power. Therefore low- be compared and analyzed. power multiplier design has been an important part in low- power VLSI system design. III. METHODS AND PERFORMANCES Fast multipliers are essential parts of digital signal processing There are number of techniques that to perform binary systems. The speed of multiplier operation is of great multiplication. In general, the choice is based upon factors importance in digital signal processing as well as in the such as speed, throughput, area, and design complexity power general purpose processors today [6]. consumption. More efficient parallel approach uses some sort In recent years, several power reduction techniques have of array or tree of full adders to sum partial products. Array been proposed for low-power digital design, including the multiplier, Wallace Tree multiplier, Dadda multipliers are reduction of supply voltage, multi threshold logic and clock most common standard multipliers. speed, the use of signed magnitude arithmetic and differential A. Array Multiplier www.ijltemas.in Page 49 International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS) Volume VI, Issue III, March 2017 | ISSN 2278-2540 Array multiplier is an efficient layout of a combinational Array multiplication can be classified into two types. These multiplier. In array multiplier, an array of identical cells are unsigned & signed. Unsigned multiplication can be done generates new partial product & accumulate of it at the same by three types - array multiplication using ripple carry adder, time. One advantage of array multiplier comes from its array multiplication using ripple carry adder & carry save regular structure .Since it is regular, it is easy to layout and adder, array multiplication using carry save adder, carry look circuit complexity is less. The design time of array multiplier ahead adder. Baugh-wooley multiplication is signed is less than tree multiplier. The second advantage of array multiplication. multiplier is its ease of design for pipeline architecture. 1.1. Array Multiplier (Using RCA) The main disadvantage of array multiplier is that it requires large number of gates, because of which area is increased. Fig.2 Array multiplier using ripple carry adder Power consumption is more as more logic gates are needed. 1.2. Array Multiplier (Using RCA&CSA) Due to this reason array multiplier is less economic. Another disadvantage of array multiplier is the worst-case delay of the Fig.3 Array multiplier using ripple carry adder & carry save adder multiplier is proportional to the width of the multiplier. The speed will be slow for wide multiplier. 1.3. Array Multiplier (Using CSA&CLA) In array multiplier, consider two binary numbers A and B, of m and n bits. There are (m×n) product terms and that are Fig.4 Array multiplier using carry save adder & carry look ahead produced in parallel by a set of (m×n) AND gates. A n×n adder array multiplier requires {n(n-2)} full adders, n half-adders and (n×n) AND gates. Also, in array multiplier worst case delay would be (2n+1)×td where td is the gate delay.[8] Delay in array multiplier is very high using ripple carry adder. Delay can be reduced if carry look ahead adder used instead of ripple carry adder in the final stage. Using these 3 methods only unsigned numbers can multiply. 1.4. Array Multiplier (Signed)/Baugh-Wooley Multiplier The array multiplier Baugh-Wooley is an efficient way for multiplying both signed and unsigned numbers. Baugh- Wooley algorithm is used in High Performance Multiplier (HPM) tree, which inherits regular and repeating structure of the array multiplier. Baugh-Wooley multiplier exhibits less delay, low power dissipation and the area occupied is also small compared to other array multipliers. The architecture of Baugh-Wooley multiplier is based on carry save algorithm[3]. Fig.1 Array multiplier www.ijltemas.in Page 50 International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS) Volume VI, Issue III, March 2017 | ISSN 2278-2540 Fig.7 Wallace multiplier C. Dadda Tree Multiplier In a parallel multiplier the partial products are generated by using array of AND gates. The main problem is the Fig.5 Baugh-wooley multiplier summation of the partial products, and it is the time taken to perform this summation which determines the maximum B. Wallace Tree Multiplier speed at which a multiplier may operate. A fast process for multiplication of two numbers was The advantage of dada multiplier is that in the Wallace developed by Wallace. Using this method, a three step process method, the partial products are reduced as soon as possible, is used to multiply two numbers; the bit products are formed, in contrast, dadda method does the minimum reduction the bit product matrix is reduced to a two row matrix where necessary at each level to perform the reduction in the same sum of the row equals the sum of bit products, and the two number of levels as required by the Wallace method resulting resulting rows are summed with a fast adder to produce a final in a design with fewer full adders and half adders. So power product [2]. consumption is less. The disadvantage of dadda method is that In the Wallace multiplier the multiplicand-multipliers are it requires a slightly wider, fast CPA and has a less regular summed up in parallel by means of a tree of carry save adders. structure than Wallace tree. A carry save adder sums up three binary numbers and The number of Halfadder and Fulladder required for Dadda produces two binary numbers. multiplier depends on N, number of bits of the operands. One advantage of Wallace tree is it has small delay. The Number of Fulladder = N^2-4.N+3 number of logic levels required to perform summation can be reduced with Wallace tree, as a result power consumption is Number of Halfadder = N-1 less. The main disadvantage of Wallace tree is complex layout Carry lookahead adder length = 2.N-2 [9]. and irregular wires. A dot diagram for an 4 by 4 Dadda multiplier is shown in A dot diagram for an 4 by 4 Wallace multiplier is shown in fig6. This architecture requires 3 Fulladder,3 Halfadder to fig4.This architecture requires 5 fulladder, 3 halfadder to sum sum the 16 partial products.The Wallace tree multiplier the 16 partial products.The Wallace tree multiplier requires a requires a carry look ahead adder which is 6 bit wide[1]. carry look ahead adder which is only 4 bit wide[1]. Fig.6 A 4×4 Wallace tree algorithm. Fig.8 A 4×4 8 Dadda tree algorithm www.ijltemas.in Page 51 International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS) Volume VI, Issue III, March 2017 | ISSN 2278-2540 D.